Somewhere this month, someone read a post, clicked the three dots, and selected an option that didn’t exist before: “Seems like AI slop.” LinkedIn quietly added that button for a subset of users and has been widening the rollout ever since, and it works on comments as well as posts. For the first time, LinkedIn members aren’t just scrolling past generic AI content, they’re actively reporting it, and that report costs the poster real reach.
The scale of the problem explains the urgency. Pangram Labs, an AI text detection company, scanned more than a million social posts and found LinkedIn to be the most AI-saturated platform it measured, with over 40% of long-form posts now fully AI-generated. LinkedIn posts made up a third of everything Pangram scanned, but nearly two-thirds of everything it flagged as AI writing. Hari Srinivasan, chief product officer of the LinkedIn ecosystem, announced the new reporting tool himself, calling the fight against AI slop “a top priority for all of us.”
The mechanics matter for anyone posting professionally. A flagged post doesn’t get deleted, it gets suppressed, the same reach penalty LinkedIn already applies to content people mark “not interested.” The person who posted it finds out privately, through their own analytics. LinkedIn is also pulling its “enhance your post” writing tool and replacing it with something narrower: a feature Srinivasan says will proofread a person’s words without changing their voice. The distinction is deliberate. LinkedIn isn’t trying to ban AI assistance. It’s trying to stop AI from doing the actual thinking.
Not everyone is convinced the system will be applied fairly. Allison Rossi, a fractional chief marketing officer, commented directly on Srinivasan’s announcement that the button essentially hands people a new way to falsely accuse posts they simply didn’t like. Jan Tegze, a talent acquisition director and LinkedIn instructor, put the deeper problem more bluntly in the same comment thread: careful, deliberate writing can end up reading as fake precisely because it’s polished, the opposite of what a hastily typed, authentic post looks like.
That tension points to what’s actually changing for anyone who posts on LinkedIn for business. According to Forbes contributor Jodie Cook, who covers AI and founder-led growth, the giveaways that get posts flagged are small and specific: an em dash dropped mid-sentence, a line that sets up an idea only to dismiss it, and a cluster of words, delve, landscape, journey, quietly, that have become shorthand for AI-generated prose. Cook argues the fix isn’t avoiding AI entirely, it’s changing where AI sits in the process: draft the actual idea yourself first, in your own words, before any AI tool touches it, then use AI strictly to catch typos and awkward phrasing rather than to shape the thinking.
The crackdown isn’t limited to posts. LinkedIn says it now blocks hundreds of thousands of automated comment attempts every single day, and the new reporting button applies to comments too. That matters because comments are one of the fastest ways a small business or a professional builds visibility and trust on the platform, and generic AI-generated replies are exactly the kind of content now being pushed toward extinction.
None of this means AI-assisted content is off the table. LinkedIn’s own position, echoed by Srinivasan, is that AI-assisted work remains welcome as long as a real idea or genuine point of view sits underneath it. What’s losing ground fast is content that’s AI-generated with nothing else added: no direct experience, no specific opinion, no detail a reader couldn’t have gotten from a thousand other identical posts in their feed. For small businesses relying on LinkedIn to reach clients, partners, and hires, that’s the actual line worth paying attention to, not whether AI touched the post, but whether anything of the poster’s own thinking survived the process.